Skip to main content
Legia

IT Support & Consulting

What Is an SLA in an IT Service Contract?

An SLA (service level agreement) is the part of a contract that sets the level of service a provider promises. Here is how to read the figures in it, from response time to availability percentage.

SEO and Content Specialist

2 min read

A team meeting in an office room

Short answer: an SLA (service level agreement) is a written commitment between an IT service provider and its customer about the level of service promised: response time, resolution time, system availability, service hours, and reporting. The figures only mean something if the way they are measured, the exclusions, and the consequences of missing them are written down as well.

The term SLA appears in almost every subscription IT service contract, from office IT support to cloud services. The NIST glossary defines it as a commitment between a service provider and one or more customers that addresses specific aspects of the service, such as responsibilities, the type of service, the expected performance level (reliability, quality, and response times), and requirements for reporting, resolution, and termination. This article explains how to read the figures in it. The full list of things to check in the contract is in the parent article, What to Check in an IT Service Contract Before Signing.

The three figures most often found in an SLA

MeasureWhat it meansWhat to confirm
Response timeThe time between a request coming in and the provider starting to handle itCounted from when: ticket creation, or the start of service hours?
Resolution timeThe time until the problem is fixed or a workaround is in placeIs it a target or a promise, and what does “resolved” mean?
AvailabilityThe percentage of time the system is usable in a periodWhich period, measured with what tool, and what is excluded?

A fast response time is not the same as a problem being fixed quickly. A provider can meet the response time just by replying to the ticket while the fix takes days. That is why the two need to be written separately.

Illustration: IT support

Reading the availability percentage

A small difference in percentage means a large difference in downtime. For a 30-day month (43,200 minutes), the numbers are:

AvailabilityDowntime per month that still meets the SLA
99%about 7 hours 12 minutes
99.5%about 3 hours 36 minutes
99.9%about 43 minutes
99.99%about 4 minutes

A real example: the Google Workspace SLA promises monthly availability of at least 99.9%. If it is missed, the customer receives 3, 7, or 15 additional days of service depending on how far the figure falls. The compensation is not automatic: the customer must file a claim within 30 days, and outages caused by force majeure or the customer’s own equipment are excluded.

Three things that make an SLA usable

  1. A measurement method you can check. Response time is recorded in the ticketing system, availability in the monitoring tool, and both appear in periodic reports.
  2. Written exclusions. Scheduled maintenance, internet provider outages, or customer-owned equipment are often excluded. Make sure the list is clear rather than decided after the fact.
  3. A procedure for when the SLA is missed. Who files, within what time, and what form the remedy takes. Without a procedure, a breach is only a note.

At Legia, every request in the IT Support & Infrastructure Management service has a ticket number, response times follow the service hours and severity levels written in the contract, and you receive a monthly report on the number of requests, problem types, and recommendations. That way, the SLA figures can be checked against the handling records.

Sources

Category IT Support & Consulting

Keep reading

Related articles

All Articles